TCL Wants To Use AI Object Recognition To Optimize Your TV’s Color Reproduction

January 11, 2019

By Janko Roettgers

CREDIT: JANKO ROETTGERS / VARIETY

TCL is giving the label “smart TV” a whole new meaning at CES: The Chinese consumer electronics company showed off a TV set that analyzes the visual content of a video, and then changes color tones accordingly.

The demo on display on the show floor Tuesday was able to identify yellow sand and sand dunes in one video example, which led it to correctly identify the video as a desert scene. Another video was identified as an autumn scene based on the recognition of fallen leaves and yellow leaves.

Each time, the TV changed the color tones accordingly, adding more warmth for the autumn scene, and making the sand dunes look more vibrant in the desert scene.

CREDIT: JANKO ROETTGERS / VARIETY

A TCL spokesperson was quick to point out that this kind of use of artificial intelligencewas just a technology demo, and not something the company is introducing with its 2019 line of TV sets. However, the technology is in many ways a successor to the company’s IPQ engine, which optimizes HDR reproduction dynamically by analyzing the visual qualities of each and every scene.

TCL isn’t the first company to tout the use of artificial intelligence for its TV sets. But while previous implementations were often constrained to voice control and content recommendations, this demo clearly showed that there may be far more advanced applications coming down the pipeline.

Read More:
https://variety.com/2019/digital/news/tcl-tv-object-recognition-1203102822/